The proposed study is a psychiatric epidemiologic investigation of a managerial and professional workforce. The overall goals are to describe the distribution of selected clinical and subclinical psychiatric conditions, and the relationships between occupational and domestic stress and these conditions. The proposed study also addresses the extent to which predispositional characteristics and psychosocial moderators influence these relationships. To meet these goals, interviews will be conducted with 2000 managers and professionals at two divisions of the Westinghouse Electric Corporation. The sample will include 1000 women and 1000 men. The interviews will include modified sections of the SADS-L, the Hopkins Symptom Checklist, and three scales focused on drinking behavior. Occupational stress measures will include a job events checklist, and scales assessing work load, job decision latitude, role conflict, role ambiguity, and the physical environment. In addition, objective stress ratings from senior level managers and objective measures of the physical environment will be obtained. Domestic stress will be assessed using the Dyadic Adjustment Scale, a measure of parental strain, and a home events checklist. Predispositional risk factors (family history of mental illness; personal history of depression or alcohol abuse; demographic factors) and psychosocial moderators (personality; family and friend support; support at work) will also be assessed. An ultimate aim of the analysis is to develop a conceptual framework based on the relationships observed which will suggest strategies for the design of future longitudinal or experimental epidemiologic research. The study is significant because it focuses on an important segment of the workforce which until now has not been studied in this manner. Two research traditions are brought together for the first time, namely, the occupational stress approach and the epidemiologic approach to studying life event stressors, particularly those occurring in the home. The fact that the design is necessarily cross-sectional means that caution will be exercised in developing measures and analyzing the data to minimize confounding and maximize the potential usefulness of the results.